AD_________________

Award Number: W81XWH-08-1-0605 TITLE: The Role of Polycomb Group Gene Bmi-1 in the Development of Prostate Cancer PRINCIPAL INVESTIGATOR: Mohammad Saleem Bhat CONTRACTING ORGANIZATION: University of Minnesota-Twin Cities Minneapolis, MN 55455 REPORT DATE: September 2011 TYPE OF REPORT: Annual PREPARED FOR: U.S. Army Medical Research and Materiel Command Fort Detrick, Maryland 21702-5012 DISTRIBUTION STATEMENT: Approved for Public Release; Distribution Unlimited The views, opinions and/or findings contained in this report are those of the author(s) and should not be construed as an official Department of the Army position, policy or decision unless so designated by other documentation.

REPORT DOCUMENTATION PAGE Form Approved

OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. 1. REPORT DATE September 2011

2. REPORT TYPEAnnual

3. DATES COVERED 1 September 2010 – 31 August 2011

4. TITLE AND SUBTITLE

5a. CONTRACT NUMBER

The Role of Polycomb Group Gene Bmi-1 in the Development of Prostate Cancer 5b. GRANT NUMBER W81XWH-08-1-0605

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S)

5d. PROJECT NUMBER

Mohammad Saleem Bhat

5e. TASK NUMBER

E-Mail: msbhat@umn.edu

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

8. PERFORMING ORGANIZATION REPORT NUMBER

University of Minnesota-Twin Cities Minneapolis, MN 55455

9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)U.S. Army Medical Research and Materiel Command Fort Detrick, Maryland 21702-5012 11. SPONSOR/MONITOR’S REPORT NUMBER(S)

12. DISTRIBUTION / AVAILABILITY STATEMENT Approved for Public Release; Distribution Unlimited 13. SUPPLEMENTARY NOTES

14. ABSTRACT We proposed to investigate the role of Bmi-1 (a member of polycomb gene family) in human prostate cancer (CaP) development. Here, we present the work accomplished during the last 5 months (after submitting the 1st annual report ) of the project. In 1st annual report we showed that Bmi-1 protein levels are highly elevated in human CaP patients and we investigated the mechanistic basis of the role of Bmi-1 in human CaP. We showed that Bmi-1-silenced CaP cells exhibit decreased proliferative and clonogenic potential. On the contrary, Bmi-1-overexpressing CaP cells exhibited the reverse. Based on the outcome of micro-array analysis, we showed that silencing of Bmi-1 caused a decrease in the cyclin D1 (Wnt target) and Bcl-2 (Sonic Hedgehog-SHH target), however an increase in p16 was observed. Conversely, overexpression of Bmi-1 exhibited the reverse effects. We generated a hypothesis that the Bmi-1 regulates the expression of Cyclin D1 and Bcl-2 by interacting with Wnt /SHH signaling in CaP cells. In the current report we provide novel findings about the transcriptional activation of Bcl-2 in CaP cells. Bcl-2 is known to be regulated by SHH signaling. However, we provide evidence showing that despite blocking SHH signaling, Bmi-1 induces the Bcl-2 expression in CaP cells suggesting the involvement of other pathway too in the regulation of Bcl-2 transcriptional activation. Another important finding of our report is that we identified those Bcl-2 acts as a novel Wnt target in CaP cells. Our investigations suggest that Bmi-1 regulates Bcl-2 through Wnt signaling. Finally, animal studies showed a significantly reduced growth in PC-3-Bmi-1-supressing cell-originated tumors than PC-3-1-Bmi-overexpressing cell-originated tumors in xenograft mouse models.

15. SUBJECT TERMS Bmi-1, Wnt Signaling, Bcl-2, TCF, Prostate Cancer

16. SECURITY CLASSIFICATION OF:

17. LIMITATION OF ABSTRACT

18. NUMBER OF PAGES

19a. NAME OF RESPONSIBLE PERSONUSAMRMC

a. REPORT U

b. ABSTRACT U

c. THIS PAGEU

UU

47

19b. TELEPHONE NUMBER (include area code)

Table of Contents

Page Introduction…………………………………………………………….………..….. 4 Body………………………………………………………………………………….. 5 Key Research Accomplishments………………………………………….…….. 8 Reportable Outcomes……………………………………………………………… 10 Conclusion…………………………………………………………………………… 16 References……………………………………………………………………………. 18 Appendices (Figure Legends & Figures) …………………………………… 23

4

This award was transferred from the University of Wisconsin to the University of Minnesota. The wok on this project was suspended at least for 3 months. This was due to the process of transfer of funding from PI’s earlier work place (University of Wisconsin) to PI’s new work place (University of Minnesota). The award was relinquished by University of Wisconsin on 01/31/2010. The University of Minnesota set up a pre-award fund (for this project) on 4/13/2010. This caused a delay in executing and completing the some of the tasks as proposed under Tasks 1-2 of the proposal. ___________________________________________________________________________ Introduction: Prostate cancer (CaP) is the most common visceral cancer diagnosed in men; it is the second leading cause of

cancer related deaths in males in the United States and the Western world (1). The lack of effective therapies for

advanced CaP reflects to a large extent, the paucity of knowledge about the molecular pathways involved in

CaP development. Thus, the identification of new predictive biomarkers will be important for improving

clinical management, leading to improved survival of patients with CaP. Such molecular targets, especially

those that are indicative of proliferation, invasiveness of the disease and survival of cancerous cells (even after

chemotherapy) will also be excellent candidate targets for staging the disease and establishing effectiveness of

therapeutic and chemopreventive intervention of CaP (2 ).

The critical pathological processes that occur during the development and progression of human CaP and are

known to confer aggressiveness to cancer cells are (i) abolishment of senescence of normal prostate epithelial

cells (ii) self-renewablity of CaP cells even after chemotherapy and radiation and (iii) dysregulated cell cycle

resulting in unchecked proliferation of cancer cells (3-4). Polycomb group (PcG) family of proteins (which form

multimeric gene-repressing complexes) have been reported to be involve in self-renewablity, cell cycle

regulation, and senescence (5-7). Bmi-1 is a transcription repressor originally identified as a c-myc cooperating

oncogene in murine lymphoma and has emerged as an important member of PcG family (8). It has been shown

to determine the proliferative potential of normal and cancer cells and is reported to be required for the self-

renewal of hematopoietic and neural stem cells (9). The human Bmi-1 is located on the short arm of

chromosome 10p13, a region known to be involved in translocations in various leukemias and rearrangements

in malignant T-cell lymphomas (9). Bmi-1 has been shown to be overexpressed in lymphomas, non–small cell

lung cancer, B-cell non-Hodgkin's lymphoma, breast cancer, colorectal cancer and nasopharyngeal carcinoma

(10). Bmi-1 was has been showed to be a useful molecular marker for predicting occurrence of myelodysplastic

5

syndrome and prognosis of the patients (11). Cellular target genes of Bmi-1 have been identified and include

ink4a and ink4b loci, encoding p16INK4A, p19ARF, and p15INK4B (12). Recently, Glinsky et al have shown that the

activation of Bmi-1 might be associated with the malignant behavior of CaP cells (13). In the current study, we

provide evidence about the over-expression of Bmi-1 in human CaP cells (in particular in highly aggressive and

androgen-independent cell types) and tissue specimens and show that this correlates with the clinical stages of

human CaP. We also show that t he over expression of Bmi-1 breaks the senescence of normal prostate

epithelial cells as well as drives proliferation of CaP cells by regulating the expression of pro-survival and

proliferation-associated genes such as Cyclin D1 and Bcl-2. We propose a role for Bmi-1 protein in CaP

development and suggest its potential use as a biomarker in the clinical management of CaP.

Body Under this section we provide information about the experimental design for tasks # 1-2 and materials and

methods used to accomplish our objectives as stated in the proposal.

Experimental Design for Specific Aim #1. We conducted the experiments to define the effect of overexpression and silencing of Bmi-1 gene in CaP cells.

For this purpose, we (a) knockdown the Bmi-1 gene by transfection of siRNA and (b) overexpressed the Bmi-1

gene by transfecting Bmi-1 construct (pbabe-Bmi-1 plasmid provided by Professor Chi Van Dang, Professor of

Cell Biology, School of Medicine, The Johns Hopkins University, Baltimore, MD) in PC3 (androgen-

independent), LNCaP (androgen-dependent), 22Rν1 (androgen-sensitive) and normal prostate epithelial cells

(PrEC) cells. We then studied the growth and viability of transfected cells in vitro by employing the MTT assay.

To investigate the effect of Bmi-1 gene on the rate of proliferation of CaP cells, we employed 3[H]thymidine

uptake assay. This assays measures the amount of 3[H]thymidine taken up by dividing cells (for DNA synthesis)

thus gives a measure of the rate of division or proliferation of cells. Bmi-1 silenced and Bmi-1 overexpressing

CaP cells were cultured in presence of 3[H]thymidine and 3[H] thymidine uptake was measured by Liquid

scintillation counter. These cells were also measured for DNA content. Since Bmi-1 was observed to increase

the proliferative potential of CaP cells and to establish that Bmi-1 indeed was a driving force for proliferating

cells, we investigated whether Bmi-1 has to potential to drive proliferation of normal prostate epithelial cells.

6

For this purpose, Bmi-1 was overexpressed in normal prostate epithelial cells (PrEC). We chose PrEC cells

because under normal culture conditions, PrEC cells are known to replicate between 3-4 cycles and after 4

cycles, these cells enter into a mode of senescence. The break of senescence in normal epithelial cells is a hall

mark of progression towards proliferation. As a control to study, another set of PrEC cells were transfected

alone vector (pbabe). Further a microarray was performed with Bmi-1 silenced LNCaP cells to understand the

mechanism of action of Bmi-1 in CaP cells. Experiments conducted under this aim provided information

whether genes involved in proliferation are regulated by Bmi-1 gene. These data were validated by western blot

analysis. We analyzed the expression level of Cyclin D1, p16 and Bcl-2 protein in CaP cells. Next we

investigated whether the overexpression generates the data contrary to what was observed in Bmi-1 silenced

cells. For this purpose Bmi-1 was overexpressed in LNCaP, PC-3 and DU145 cells by transfecting pbabe-Bmi-1

plasmid. Cell lysates prepared from these cells were analyzed for Cyclin D1, Bcl-2 and p16 proteins by

employing western blot analysis. To understand the mechanism through which Bmi-1 regulates Cyclin D1, we

carried out experiments on critical pathways which are already know to be associated with Cyclin D1

expression. This includes Wnt/β-catenin signaling pathway. We asked whether Bmi-1 has any association with

Wnt/β-catenin signaling (which is itself reported to control Cyclin D1). Interestingly, we found that Bmi-1

overexpression causes an increase in the transcriptional activation of TCF-responsive element (a bio-marker of

Wnt signaling) in CaP cells. Since Bcl-2 was observed to be modulated by Bmi-1, we investigated if Bmi-1 has

any association with sonic hedgehog (SHH) pathway that is very well know to regulate Bcl-2. For this purpose

we determined the expression level of Bcl-2 in Bmi-1-overexpressing and Bmi-1-silenced CaP cells in presence

of Cyclopamine, a SHH pathway inhibitor. We also tested if re-introduction of Bmi-1 would restore the Bcl-2

levels in CaP cells pre-treated with cyclopamine (SHH inhibitor). Further, we investigated an association of tcf

and Bcl-2 in CaP cells. We investigated the mechanism through which Bmi-1 drives the Tcf/Bcl-2 signaling in

CaP cells. Finally, animal studies showed a significant lower tumor growth in PC-3-empty vector and PC-3-

Bmi-1-supressing cell-originated tumors than PC-3-Bmi-1-overexpressing c ell originated tumors in athymic

mice

7

Material and Methods:

Cell Lines: Normal prostate epithelial cell (PrEC) was procured from Cambrex. Virally transformed prostate

epithelia cells (RWPE-1), PC-3, 22Rν1, DU-145 and LNCaP cancer cells were obtained from ATCC

(Manassas, VA). Cells were cultured in appropriate media and were kept in a humidified atmosphere of 95% air

and 5% CO2 in an incubator at 37 oC.

Plasmids and siRNA: The pbabe-Bmi-1 plasmid was a kind gift from Dr. Chi V. Dang (The John Hopkins

University, Baltimore, MD). pbabe vector was purchased from Addgene Inc. (Cambridge, MA). Bmi-1-siRNA

and scrambled siRNA were commercially purchased from Dharmacon (Lafayette, CO). Vector-based shRNA

plasmid pGeneCLIP and pGeneCLIP-Bmi-1-shRNA were procured from SA Biosciences Corporation

(Fredrick, MD)

Transfections. For siRNA transfection studies, CaP cells were plated at a density of 1 x 106 cells per well in 6-

well plates and incubated for 24, 36 a nd 48 h i n complete medium. Using Amaxa nucleofactor kit

(Gaithersburg, MD), cells were transfected with siRNAs i.e., non silencing siRNA (100 nM) and Bmi-1 siRNA

(100 nM). Cells were harvested after 24, 36 and 48 h and analyzed for expression of Bmi-1. For pbabe-Bmi-1

plasmid transfection studies, CaP cells (1 x 106 cells per well) were transfected with 1-2 µg of the Bmi-1

construct. For controls, the same amount of empty vector, pbabe and GFP vector (as positive control for

transfection) were also transfected. Cells were harvested after 24, 36 and 48 h and analyzed for expression of

Bmi-1. For overexpressing Bmi-1 stably, retroviral transfection of CaP cells was performed.

Transcriptional activity of TCF and Bcl-2: pGL3-Bcl-2 was procured from Dr. Vladmier Speiglaman

(Department of Dermatology, University of Wisconsin, Madison, WI) pTK-TCF-Luc (TopFlash & FopFlash)

was procured from Upstate Laboratories (Lake Placid, NY). Cells were transfected with the plasmids (200

ng/well) for 24 h. Renilla luciferase (20 ng/well, pRL-TK; Promega, Madison, WI) was used as an internal

control. In addition, for controls, the same amount of empty vectors, were transfected in cells. The cells were

then harvested and transcriptional activity was measured in terms of luciferase activity in quadruplicates by

using dual-luciferase reporter assay system (Promega, Madison, WI).

8

Western blot analysis. Cell lysates were prepared in cold lysis buffer [(0.05 mmol/L Tris-HCl, 0.15 mmol/L

NaCl, 1 mole/L EGTA, 1 mol/L EDTA, 20 mmol/L NaF, 100 mmol/L Na3VO4, 0.5% NP-40, 1% Triton X-100,

1 mol/L phenyl methylsulfonyl flouride (pH 7.4)] with freshly added protease inhibitor cocktail (Protease

Inhibitor Cocktail Set III, Calbiochem, La Jolla, CA). The lysate was collected, cleared by centrifugation,

supernatant aliquoted and stored at -80 °C. The protein content in the lysates was measured by BCA protein

assay (Pierce, Rockford, IL), as per the manufacturers’ protocol. For Western blot analysis, 40 µg protein was

resolved over 12% Tris-glycine polyacrylamide gels (Novex, Carlsbad, CA) under non-reduced conditions,

transferred onto nitrocellulose membranes and subsequently incubated in blocking buffer (5% nonfat dry

milk/1% Tween 20; in 20mmol/L TBS, pH 7.6) for 2 hours. The blots were incubated with appropriate primary

(human reactive Bmi-1, Cyclin D1, Bcl-2 and p16), washed and incubated with appropriate secondary HRP-

conjugated antibody (Amersham Biosciences, Piscataway, NJ). The blots were detected with

chemiluminescence (ECL kit, Amersham Biosciences, Piscataway, NJ) and autoradiography, using XAR-5 film

(Eastman Kodak, Rochester, NY). Equal loading of protein was confirmed by stripping the blots and reprobing

with β-actin (Sigma, St.Louis, MO).

Key Research Accomplishments: We proposed three sub-aims under specific aim # 1. Following is the list of work completed under 1-17 months

proposed time as described in the Statement of Work:

Proposed Task 1. To establish the involvement of Bmi-1 in the proliferation of human CaP cells. Status : Submitted as 1st annual report

Proposed Task 2. To evaluate the effect of Bmi-1 overexpression and silencing on SHH and Wnt/β-catenin signaling pathways and to define the underlying mechanism of SHH-Bmi-1-β-catenin interaction at transcriptional and translational level in human CaP cells. :

Status: Results generated under Sub aim# A of this task were submitted as Mid-term report; Work proposed under Sub-aim B of this task is underway Proposed Task 3 (A)To Studies in athymic nude mouse xenograft model will be conducted (a) to analyze

the consequences of Bmi-1 overexpression and silencing on tumorigenicity of human CaP cells under in vivo situation and (b) to evaluate the effect of Bmi-1 siRNA in combination with inhibitors of SHH and β-catenin signaling on the growth of tumors.

Status : Experiments are underway and following tasks have been completed:

9

(i). Stable Bmi-1-overexpressing CaP Cell Generation: Stable CaP cells lines overexpressing Bmi-1 have

been generated. For this purpose PC-3 cells were stably transfected with pbabe-Bmi-1. The transfections

were performed by Lipofectamine method. Bmi-1 overexpressing PC-3 clones were selected in presence of

puromycin. The selection of Bmi-1-overexpressing PC-3 cells under puromycin continued for 4 weeks.

Cells were tested for Bmi-1 overexpression. Among 24 clones generated, we selected 3 c lones those

exhibited the highest degree of Bmi-1 expression level (Figure 15A).

(ii). Stable Bmi-1-silenced CaP Cell Generation by vector-based shRNA: Stable CaP cells lines exhibiting

least or no Bmi-1 expression levels have been generated. For this purpose PC-3 cells were stably transfected

with vector-based shRNA plasmid, pGeneCLIP-Bmi-1-shRNA. The shRNA plasmids are designed using an

experimentally validated algorithm. These constructs specifically knock down the expression of specific

genes by RNA interference and allow for enrichment or selection of transfected cells. Each vector expresses

a short hairpin RNA, or shRNA, under control of the U1 promoter and neomycin gene. Neomycin resistance

permits selection of stably transfected cells. The ability to select or track and enrich shRNA-expressing cells

brings RNA interference to cell lines with lower transfection efficiencies. Unlike siRNA, plasmid-based

shRNA also provide a renewable source of RNA interference reagent. The transfections were performed by

Lipofectamine method. PC-3 cells were selected in presence of neomycin analogue G418 (300 µg/ml). The

selection of Bmi-1-overexpressing PC-3 cells under G418 continued for 4 weeks. Cells were tested for Bmi-

1 expression. Among 20 clones generated, we selected 3 clones those exhibited the least or no Bmi-1

expression level (Figure 15A).

(iii). Use of stable CaP cells for tumor studies in xenograft mouse models. Athymic (nu/nu) male nude mice

(HarlanTek, Madsion, WI), were housed under pathogen-free conditions with a 12 h l ight/12 h dark schedule

and fed with an autoclaved diet ad libitum. Empty vector transfected PC-3 clones (PC-3), Bmi-1 over

expressing PC-3 clones (BO) and Bmi-1 suppressing PC-3 clones (BS) were selected for tumor studies. 1 x 106

cells mixed with 50 μL RPMI + 100 μL Matrigel (BD, Franklin Lakes, NJ) were injected subcutaneously in the

right flanks of each mouse, with 10 animals in each group. Body weights were recorded five days weekly

throughout the study. Tumor volumes were recorded three times/ week. The tumor volume was calculated by

10

the formula 0.5238 x L1 x L2 x L2, where L1 is the long diameter and L2 is the short diameter of the tumor. All

animals of group 1 and group 2 were sacrificed when tumors reached a pre-set endpoint volume of 1,000 mm3.

All procedures conducted were in accordance with the IACUC guidelines.

Statistical analysis. All measures were summarized as means ± SE. Measures were examined for the

appropriateness of a normality assumption by density estimation (data not shown). Associations of categoric

variables were evaluated using the Fisher’s exact test. Sample correlations were estimated using Spearman’s

rank correlation. All tests were two-sided and conducted at the alpha = 0.05 significance level. All statistical

analyses were performed with the S-plus, Professional Version 6.2 (Insightful Corp., Seattle, WA) software.

Reportable Outcome Bmi-1 protein expression in normal and CaP cells: As an attempt towards identifying the expression of Bmi-1

in CaP progression, we first measured protein expression levels by immunoblot analysis in several human

prostate carcinoma cell lines, LNCaP, DU-145 and PC-3, and compared them to NHPE and RWPE-1 cells.

Among three cell lines used, LNCaP is androgen-sensitive whereas DU-145 and PC-3 are androgen-

independent. The choice of these cells was based on t he fact that 80% CaP patients present with androgen-

dependent disease at the time of diagnosis which later transforms into more aggressive, androgen-independent

disease (14-15 and references therein). As shown in Figure 1A, all CaP cell lines exhibited a higher expression

of Bmi-1 protein than in normal prostate epithelial cells. When the protein expression of Bmi-1 was compared

among three cancer lines, based on the densitometric analysis of the immunoblots, highly aggressive PC-3 cells

and DU145 exhibited 2.5-fold (p<0.001) higher expression than in LNCaP cells. These data suggest a

possibility that expression of Bmi-1 protein may be correlated with disease progression and may play a role in

aggressiveness of human CaP.

Immunohistochemical analysis of Bmi-1 protein in normal and CaP specimens: In the next series of

experiments, we used immunoperoxidase to determine Bmi-1 protein expression in specimens of age-matched

normal and CaP representing all tumor stages. In the first experiment, a total of 80 samples were obtained. The

staining intensity in tissue specimens were scored on a scale of 0-3. The staining pattern of Bmi-1 protein was

compared in grade 1, grade 2 and grade 3 CaP specimens. The mean Bmi-1 expression was 1.5 ± 0.15 (mean ±

11

S.E; n = 25) in grade 1 specimens, 2.5 ± 0.20 (mean ± S.E; n = 25) in grade 2 specimens, and 2.9 ± 0.15 (mean

± S.E; n = 30 ) in grade 3 s pecimens (Fig. 1B). These data show a p rogressive increase of Bmi-1 protein

expression corresponding with increasing tumor grade in human CaP (Figure 1B).

Bmi-1 knockdown and Bmi-1 overexpression: Bmi-1 was knocked-down in CaP cells by employing siRNA

technique. Cells were analyzed for Bmi-1 protein level at 24 and 36 h post-transfection. Bmi-1 protein levels

was observed to be highly reduced at 36 h pos t-transfection (Figure 2A). This time point was selected for

further experiments and biochemical assay utilizing Bmi-1 siRNA transfections. Overexpression of Bmi-1 was

achieved in CaP cells by employing transfecting pbabe-Bmi-1 plasmid in CaP cells. Cells were analyzed for

Bmi-1 protein level at 24 and 36 h post-transfection. Bmi-1 protein level was observed to be highly increased at

36 h post-transfection in CaP cells. The data representing Bmi-1 protein at 36h is presented in Figure 2B. This

time point was selected for further experiments and biochemical assay utilizing Bmi-1 siRNA transfections.

Effect of Bmi-1 knockdown and Bmi-1-overexpression on the growth of CaP cells: Next we investigated

effect of Bmi-1 on the growth and viability of CaP cells by employing MTT assay. LNCaP cells are known to

duplicate under culture conditions from 48-72 h. Similarly DU145 and PC-3 cells duplication takes 24 h under

culture conditions. Culture dishes containing LNCaP, DU145 and PC-3 cells become confluent between 48-72

h. It is noteworthy that Bmi-1-silenced CaP cells did only grow between 50-65% even after 72 h pos t-

transfection (Figure 3A). On the contrary, Bmi-1-overexpressing CaP cells exhibited significantly increased

growth (Figure 3B). At 36 h pos t transfection control LNCaP cells displayed 35% growth while as Bmi-1

overexpressing LNCaP cells displayed 60 % growth. Similarly Bmi-1-overexpressing DU145 and PC-3 cells

exhibited 75-100 % cell confluency at 36 h pos t-transfection as compared to control which exhibited 50%

confluency (Figure 3B). These data suggest the importance of Bmi-1 in the growth of CaP cells.

Effect of Bmi-1 knockdown and Bmi-1-overexpression on of proliferation of CaP cells. We investigated

whether Bmi-1 regulates the proliferation process of CaP cells. We employed a two-way approach where Bmi-1

was either knocked down or overexpressed in CaP cells and such CaP cells were later assessed for their

proliferative and clonogenic potential by employing 3[H]thymidine uptake assay. Firstly, LNCaP, PC-3 and

DU145 cells were transfected with Bmi1-siRNA (100 nM). To investigate the effect of Bmi-1 gene suppression

12

on the rate of proliferation of LNCaP, DU145 and PC-3 cells, we performed 3[H]thymidine uptake assay.

Suppression of Bmi1-1 expression resulted in the decreased rate of proliferation of CaP cells (Figure 4A). Bmi-

1-overexpressing CaP cells displayed significantly increased rate of proliferation (Figure 4B).

Colony formation assay is used as a marker for proliferation of cells under ex-vivo conditions. Next, we

investigated the effect of suppression of Bmi-1 on the clonogenic potential of CaP cells in ex-vivo conditions.

For this purpose we employed a soft agar colony assay and assessed the clonogenic potential of CaP cells

transfected with Bmi-1 siRNA and pbabe-Bmi-1 plasmid. Bmi-1 knockdown resulted in a significant reduction

in the number of colonies formed by CaP cells as compared to control siRNA-treated cells (Figure 5A).

However, Bmi-1-overexpressing CaP cells exhibited increased clonogenicity potential as is evident from an

increase in the average number of colonies formed by Bmi-1-overexpressing cells (Figure 5B). These data

suggested that Bmi-1 confers proliferative attributes to the CaP cells.

Effect of Bmi-1 gene knockdown on CaP-associated genes: Next, we investigated the mechanism through

which Bmi-1 controls the proliferation and survival of CaP cells. We performed a focused microarray analysis

of 288 well-characterized proliferation and survival-associated genes in Bmi-1 suppressed-CaP cells. The list of

genes that were observed to exhibit changes in their expression pattern in response to Bmi-1 knockdown in CaP

cells is presented in Table 1. Most notably, we observed that suppression of Bmi-1 gene in CaP cells caused a

significant reduction (> 95%) in the expression of Cyclin D1, Bcl-2, urokinase plasminogen activator (uPA),

matrix metalloproteinase (MMP)-9 and nuclear factor kappa B (NFκB) (Table 1). We also observed an

increased expression of p16, p15 and TIMP-3 (Table 1). Since Cyclin D1 and Bcl-2 were highly responsive to

Bmi-1 gene suppression; we selected these genes for further biochemical studies.

Effect of Bmi-1 gene knockdown and Bmi-1-overexpression on Cyclin-D1 and Bcl-2 Levels: Increased

Cyclin-D1 activity and Bcl-2 are considered important for the increased proliferation and survival of cancerous

cells (16-21). Next, we analyzed the effect of Bmi-1 gene-suppression on the expression of cyclin D1 and Bcl-2

in LNCaP, DU145 and PC-3 cells. Bmi-1 knockdown caused a decrease in the expression level of Cyclin D1

and Bcl-2 protein in CaP cells (Figure 6A). Next we determined the effect of Bmi-1 overexpression on t he

expression levels of cyclin D1 and Bcl-2. CaP cells transfected with Bmi-1 construct exhibited a significant

13

increase in the expression level of cyclin D1 and Bcl-2 protein (Figure 6B). These data were consistent with

microarray data suggesting a possible association between the Bmi-1, Cyclin D1 and Bcl-2 during progression

of human CaP.

Effect of Bmi-1-overexpression on transcriptional activation of TCF-responsive element: Cyclin D1

expression has been reported to be regulated by Wnt signaling (22). Since we observed that Bmi-1 also regulate

Cyclin D1 expression, we asked whether there is any association between Bmi-1 and Wnt signaling. Next we

investigated effect of Bmi-1 overexpression on Wnt signaling by evaluating the transcriptional activation of Tcf-

responsive element (marker of Wnt signaling) by employing luciferase reporter assay. It is noteworthy that

Bmi-1 over-expression caused an increase in the transcriptional activation of TCF-responsive element

suggesting that Bmi-1-induced Cyclin D1 expression might be through the activation of Wnt signaling (Figure

7). This is the first report where Bmi-1 is shown to regulate Wnt signaling. To fully understand the association

between Bmi-1 and Wnt signaling, the work is underway.

Effect of Bmi-1 overexpression on the replicative life of normal prostate epithelial cells (PrEC): Transfection

of pbabe-Bmi-1 plasmid significantly increased the replicative life of PrEC cells upto 8 passages which

generally enters into a stage of senescence after 4-5 divisions (Figure 8).These data suggest that Bmi-1 possess

the potential to drive normal cells towards proliferation.

Effect of Sonic hedgehog (SHH) signaling inhibition on Bmi-1-silenced and Bmi-1 overexpressing CaP

cells: Keeping in view that (1) modulations in Bmi-1 expression cause modulations in the expression levels of

anti-apoptotic protein Bcl-2 as shown in Figure 6, and (2) the transcriptional activation of Bcl-2 is known to be

regulated by sonic hedgehog signaling, we next determined the effect of cyclopamine (SHH signaling inhibitor)

treatment on the growth and viability of LNCaP and PC-3 cells exhibiting varied expression levels of Bmi-1. To

achieve our objective, Bmi-1-suppressed and Bmi-1 overexpressing CaPs were treated with Cyclopamine for 12

h. A s is shown in Figure 9A, Cyclopamine treatment was observed to cause 35% reduction of viability of

control CaP cells (transfected with scrambled siRNA alone). Bmi-1-deficient CaP cells were significantly

responsive to cyclopamine treatment and 90% reduction in the viability of Bmi-1-suppressed cells was observed

(Figure 9A). On the contrary, Bmi-1-overexpressing CaP cells were non-responsive to Cyclopamine treatment

14

(Figure 9B). Bmi-1 over-expression was observed to abrogate the effect of Cyclopamine treatment in CaP cells

(Figure 9B). When tested for Bcl-2 protein, Bmi-1 overexpressing CaP cells were observed to exhibit elevated

levels and the reverse was observed in Bmi-1 deficient LNCaP cells when were treated with cyclopamine

(Figure 10A-B). These data suggest that Bmi-1 confers the survivability characteristics to CaP cells by inducing

the expression levels Bcl-2 (the anti-apoptotic protein) post chemotherapeutic treatment (Figures 9 & 10). These

data provide evidence that in chemoresistant CaP cells, Bcl-2 activation is SHH-independent.

Analysis of Promoter Region of Bcl-2 gene for TCF- transcriptional factor binding sites: Since we observed

that Bmi-1 induces the activation of (1) TCF, transcriptional factor and (2) Bcl-2, the anti-apoptotic factor, we

investigated the possibility of interaction among Bmi-1, Bcl-2 and TCF. We investigated possible binding sites

on the promoter region of Bcl-2 gene by employing web-based TESS analysis. Interestingly, Bcl-2 promoter

region exhibited multiple sites where TCF transcriptional factor possess the affinity to bind (Figure 11). These

data suggested that Bcl-2 is a target of Wnt signaling.

Effect of TCF-knockdown on the transcriptional activation of Bcl-2 in CaP cells: On the basis of our

observations that (1) Bmi-1 induces Bcl-2-independent of SHH signaling activation ,(2) Bmi-1 induces TCF

transcriptional activation, and (3) that Bcl-2 promoter region has multiple sites for TCF binding, we next asked

whether Bcl-2 is itself a target of Bmi-1-induced Wnt signaling. For this reason, TCF expression was knocked-

down in LNCaP and PC-3 cells by employing TCF-specific shRNA. As compared with control CaP cells

(transfected with scrambled shRNA alone), the transcriptional activation of Bcl-2 was significantly reduced in

CaP cells-deficient of TCF (Figure 12). These data suggest that Bcl-2 expression is regulated by Wnt signaling

in CaP cells.

Effect of Bmi-1 introduction on Bcl-2 promoter activity in TCF-silenced LNCaP cells: We asked if Bmi-1

introduction in Tcf-silenced cells could restore Bcl-2 transcriptional activation in cells. For this purpose LNCaP

cells were transfected with Tcf-dominant negative vector to downregulate the expression of Bmi-1. These cells

were transfected with Bcl-2-luc construct and Bcl-2 promoter activity was measured by reporter assay. As

expected, Tcf-dn LNCaP cells exhibited reduced Bcl-2 promoter activity. However, when a set of LNCaP cells

(cultured under similar conditions) exhibiting dominant negative Tcf were transfected again with Bmi-1-

15

overexpressing plasmid, Bcl-2 promoter activity was observed to be significantly restored. These data suggest

that Bmi-1 has the potential to induce Bcl-2 expression in CaP cells. Further, when tested, these cells also

exhibit restored Tcf-transcriptional activation.

Effect of Bmi-1 introduction on Bcl-2 promoter activity in LNCaP cells pretreated with SHH inhibitor: We

asked if Bmi-1 introduction in LNCaP cells pretreated with SHH inhibitor, Cyclopamine (that reduces Bcl-2

levels) could restore Bcl-2 transcriptional activation in these cells. For this purpose LNCaP cells were treated

with Cyclopamine for 24h to inhibit SHH pathway activation thus downregulating the levels of the target

protein Bcl-2 (SHH target). These cells were transfected with Bcl-2-luc construct and Bcl-2 promoter activity

was measured by reporter assay. As expected, Cyclopamine-treated LNCaP cells exhibited reduced Bcl-2

promoter activity. However, when a set of cyclopamine-treated LNCaP cells (cultured under similar conditions)

were transfected again with Bmi-1-overexpressing plasmid, Bcl-2 promoter activity was observed to be

significantly restored. These data suggest that Bmi-1 has the potential to induce Bcl-2 expression in CaP cells

even under extreme conditions (Bcl-2 ablation in this case). We hypothesize that this might be one of the

potential mechanisms through Bmi-1 confers survivability to cancerous cells undergoing chemotherapy. To

summarize, we show that Bmi-1 confers the survivability to chemoresistant CaP cells and identified Bcl-2 as a

target of Wnt signaling in CaP cells.

Investigating the significance of Bmi-1 in tumorigenic growth of prostatic epithelial cells in a xenograft mouse model: To achieve this objective we performed two protocols mentioned as following: (a) Effect of Bmi-1 overexpression on tumorigenicity of human prostate cancer cell-derived tumor implanted

in a xenograft mouse model:

Since Bmi-1 was observed to be involved in the growth of CaP in vitro, we next determined whether the growth

activities of Bmi-1 (observed in vitro) would translate into an in vivo xenografts model. Implantation of empty

vector transfected PC-3 cells (PC-3) and Bmi-1-overexpressing PC-3 (termed as PC-3-BO in Figure 16) cells

onto nude mice produced visible tumors in mice with a mean latent period of 7 days. The average volume of

PC-3 and PC-3-BO-tumors in mice increased as a function of time and crossed a preset end point of 1000 mm3

at 49th and 35th days post-inoculation. At 49th and 35th days (of post inoculation), mice implanted with PC-3-

16

derived and PC-3-BO-derived tumors exhibited average tumor volumes of 1076 and 1023 mm3 respectively

(Figure 16A and C).

(b) Effect of Bmi-1 silencing on tumorigenicity of human prostate cancer cell-derived tumor implanted in a

xenograft mouse model:

Implantation of empty vector transfected PC-3 cells (PC-3) and PC-3-Bmi-1-supressing cells (termed as PC-3-

BS in Figure 17) onto nude mice produced visible tumors in mice with a mean latent period of 7 days. The

average volume of PC-3 and PC-3-BS-tumors in non-treated mice increased as a function of time and crossed a

preset end point of 1000 mm3 at 49th and 70th days post-inoculation. At 49th and 70th days (of post inoculation),

mice implanted with PC-3-derived and PC-3-BS-derived tumors exhibited average tumor volumes of 1176 and

869 mm3 respectively Figure 17A and B).

Conclusion

Recent experimental observations documented an increased Bmi-1 expression in human non-small-cell lung

cancer, human breast carcinomas, and established breast cancer cell lines, suggesting that an oncogenic role of

Bmi-1 activation may be extended beyond leukemia and, perhaps, may affect progression of the epithelial

malignancies as well (8-13, 23-24). Over expression of Bmi-1 is reported to confer invasive potential to glioma

and breast cancer cells, and cause malignant phenotype in rat and human cancer cells, however, the role of Bmi-

1 in human CaP metastasis is yet to be elucidated (8-13, 23-24). In the present study, we observed that Bmi-1

gene controls the invasiveness and growth of human CaP cells under in vitro and in vivo conditions through the

regulation of Cyclin D1 and Bcl-2. To our knowledge, this report is the first demonstration that Bmi-1 regulates

Cyclin D1 and Bcl-2 in human CaP cells.

In our preliminary studies (data generated in the first year of the project), we show that Bmi-1 protein levels are

elevated in CaP patients if high grade tumor. Further, we also show that overexpression of Bmi-1 gene increases

the rate of proliferation and invasion of CaP cells and its suppression reverses this effect. We provide evidence

that suppression of Bmi-1 gene reduces the proliferative and clonogenic potential of human CaP cells. On the

contrary, overexpression of Bmi-1 was observed to increase the clonogenic potential of human CaP cells.

17

Cyclin D1 is a critical protein that is required for cell division and proliferation. Cyclin D1 levels have been

reported to be increased during several cancer types (25-33). CaP patients have been shown to exhibit increased

Cyclin D1 levels (34-35), however the mechanism through which Cyclin D1 levels are altered during cancer

development is not fully understood. In this study, we provide evidence that Cyclin D1 is regulated by Bmi-1 in

CaP cells. It is noteworthy that CaP cells deficient in Bmi-1 exhibited decreased Cyclin D1 levels. However,

Bmi-1-overexpressing CaP cells displayed an increase in Cyclin D1 protein levels suggesting significance of

Bmi-1 protein for Cyclin D1 expression in CaP cells.

To understand the mechanism through which Bmi-1 regulates Cyclin D1, we studied critical pathways which

are known to be associated with Cyclin D1 expression. This includes Wnt/β-catenin signaling pathway. We

asked whether Bmi-1 has any association with Wnt/β-catenin signaling (which is itself reported to control

Cyclin D1). Interestingly, we found that Bmi-1 overexpression causes an increase in the transcriptional

activation of TCF-responsive element (a bio-marker of Wnt signaling) in CaP cells. These data are highly

significant. This finding is novel and is the first report showing that Bmi-1 regulates Wnt signaling in CaP cells.

Wnt signaling is report to be involved in the proliferation and chemoresistance of CaP cells. Further work to

understand the mechanistic role in Wnt signaling in CaP cells is underway and will be completed by the end of

2nd year of the proposed project.

Bcl-2 protein is known to play an important role in the survival of cancer cells by conferring anti-apoptotic

potential to cells (18-21). Bcl-2 levels are reported to be high in cancer cells including CaP cells (36). In the

current study, we provide evidence that Bmi-1 is associated with expression of Bcl-2 in CaP cells. Bmi-1-

deficient CaP cells exhibited decreased Bcl-2 levels while as Bmi-1-overexpressing CaP cells exhibited

increased Bcl-2 levels. As evident from reports which suggest that Bmi-1 confers renewability or stemness

characteristics to cancer cells and Bcl-2 confers survivability characteristics to cancer cells, our data showing

association between Bmi-1 and Bcl-2 is highly significant. However, it would be important to fully understand

the mechanism through which Bmi-1 regulates Bcl-2 in CaP cells. To understand the mechanism through

which (1) Bmi-1-induces Wnt signaling, and (2) Wnt-signaling regulate Bcl-2 expression in CaP cells, we have

18

planned experiments and some of experiments are in progress. We provided compelling evidence showing that

Bmi-1 play an important role in growth of tumor cells in vivo. Targeting Bmi-1 by gene therapy could be a

future option for clinicians to sensitize non-treatable/chemoresistant prostatic tumors for chemotherapy. Bmi-1

could be developed as molecular target for gene therapy and chemotherapeutic agents.

19

References

1. Siegel R, Ward E, Brawley O, Jemal A. Cancer statistics. CA Cancer J Clin 2011; 61: 212-36.

2. Bickers B, Aukim-Hastie C. New molecular biomarkers for the prognosis and management of prostate

cancer--the post PSA era. Anticancer Res. 2009 Aug; 29(8):3289-98.

3. Altieri DC, Languino LR, Lian JB, Stein JL, Leav I, van Wijnen AJ, Jiang Z, Stein GS. Prostate cancer

regulatory networks. J Cell Biochem. 2009 Aug 1;107(5):845-52..

4. Kasper S. Identification, characterization, and biological relevance of prostate cancer stem cells from

clinical specimens. Urol Oncol. 2009 May-Jun;27(3):301-3.

5. Müller J, Verrijzer P. Biochemical mechanisms of gene regulation by polycomb group protein

complexes. Curr Opin Genet Dev. 2009 Apr;19(2):150-8.

6. Schuettengruber B, Cavalli G. Recruitment of polycomb group complexes and their role in the dynamic

regulation of cell fate choice. Development. 2009 Nov;136(21):3531-42.

7. Bracken AP, Helin K. Polycomb group proteins: navigators of lineage pathways led astray in cancer.

Nat Rev Cancer. 2009 Nov;9(11):773-84.

8. Kang MK, Kim RH, Kim SJ, Yip FK, Shin KH, Dimri GP, Christensen R, Han T, Park NH. Elevated

Bmi-1 expression is associated with dysplastic cell transformation during oral carcinogenesis and is

required for cancer cell replication and survival. Br J Cancer. 2007 Jan 15; 96(1):126-33.

9. Raaphorst FM. Self-renewal of hematopoietic and leukemic stem cells: a central role for the Polycomb-

group gene Bmi-1. Trends Immunol. 2003 Oct;24(10):522-4.

10. van Kemenade FJ, Raaphorst FM, Blokzijl T, Fieret E, Hamer KM, Satijn DP, Otte AP, Meijer CJ.

Coexpression of Bmi-1 and EZH2 polycomb-group proteins is associated with cycling cells and degree

of malignancy in B-cell non-Hodgkin lymphoma. Blood. 2001 Jun 15;97(12):3896-901.

11. Mihara K, Chowdhury M, Nakaju N, Hidani S, Ihara A, Hyodo H, Yasunaga S, Takihara Y, Kimura A.

Bmi-1 is useful as a novel molecular marker for predicting progression of myelodysplastic syndrome

and patient prognosis. Blood. 2006 Jan 1;107(1):305-8.

20

12. Kim JH, Yoon SY, Kim CN, Joo JH, Moon SK, Choe IS, Choe YK, Kim JW. The Bmi-1 oncoprotein is

overexpressed in human colorectal cancer and correlates with the reduced p16INK4a/p14ARF proteins.

Cancer Lett. 2004 Jan 20;203(2):217-24.

13. Glinsky GV, Berezovska O, Glinskii AB. Microarray analysis identifies a death-from-cancer signature

predicting therapy failure in patients with multiple types of cancer. J Clin Invest. 2005 Jun;115(6):1503-

21.

14. Bonkhoff H, Berges R. From pathogenesis to prevention of castration resistant prostate cancer. Prostate.

2009 Sep 16.

15. Brooke GN, Bevan CL. The role of androgen receptor mutations in prostate cancer progression. Curr

Genomics. 2009 Mar;10(1):18-25.

16. Knudsen KE. The cyclin D1b splice variant: an old oncogene learns new tricks. Cell Div. 2006 Jul

24;1:15.

17. Fu M, Wang C, Li Z, Sakamaki T, Pestell RG. Minireview: Cyclin D1: normal and abnormal functions.

Endocrinology. 2004 Dec;145(12):5439-47.

18. Patel MP, Masood A, Patel PS, Chanan-Khan AA. Targeting the Bcl-2. Curr Opin Oncol. 2009

Nov;21(6):516-23. PubMed PMID: 19730103.

19. Cotter TG. Apoptosis and cancer: the genesis of a research field. Nat Rev Cancer. 2009 Jul;9(7):501-7.

Review. PubMed PMID: 19550425.

20. Kang MH, Reynolds CP. Bcl-2 inhibitors: targeting mitochondrial apoptotic pathways in cancer therapy.

Clin Cancer Res. 2009 Feb 15;15(4):1126-32. Review.ubMed PMID: 19228717.

21. Brunelle JK, Letai A. Control of mitochondrial apoptosis by the Bcl-2 family. J Cell Sci. 2009 Feb

15;122(Pt 4):437-41.

22. Rohrs S, Kutzner N, Vlad A, Grunwald T, Ziegler S, Müller O. Chronological expression of Wnt target

genes Ccnd1, Myc, Cdkn1a, Tfrc, Plf1 and Ramp3. Cell Biol Int. 2009 Feb 11.

23. Song LB, Zeng MS, Liao WT, Zhang L, Mo HY, Liu WL, Shao JY, Wu QL, Li MZ, Xia YF, Fu LW,

Huang WL, Dimri GP, Band V, Zeng YX. Bmi-1 is a n ovel molecular marker of nasopharyngeal

21

carcinoma progression and immortalizes primary human nasopharyngeal epithelial cells. Cancer Res.

2006 Jun 15;66(12):6225-32.

24. Dimri GP, Martinez JL, Jacobs JJ, Keblusek P, Itahana K, Van Lohuizen M, Campisi J, Wazer DE,

Band V. The Bmi-1 oncogene induces telomerase activity and immortalizes human mammary

epithelial cells. Cancer Res. 2002 Aug 15;62(16):4736-45.

25. Rodriguez-Puebla ML, Robles AI, Conti CJ. ras activity and cyclin D1 expression: an essential

mechanism of mouse skin tumor development. Mol Carcinog. 1999 Jan;24(1):1-6.

26. Barnes DM, Gillett CE. Cyclin D1 in breast cancer. Breast Cancer Res Treat. 1998;52(1-3):1-15.

27. Della Ragione F, Borriello A, Della Pietra V, Cucciolla V, Oliva A, Barbarisi A, Iolascon A, Zappia V.

Cell division cycle alterations and human tumors. Adv Exp Med Biol. 1999;472:73-88.

28. Sandhu C, Slingerland J. Deregulation of the cell cycle in cancer. Cancer Detect Prev. 2000;24(2):107-

18.

29. Diehl JA. Cycling to cancer with cyclin D1. Cancer Biol Ther. 2002 May-Jun;1(3):226-31.

30. Lamb J, Ewen ME. Cyclin D1 and molecular chaperones: implications for tumorigenesis. Cell Cycle.

2003 Nov-Dec;2(6):525-7. Review. PubMed PMID:14504466.

31. Rowlands TM, Pechenkina IV, Hatsell S, Cowin P. Beta-catenin and cyclin D1:connecting development

to breast cancer. Cell Cycle. 2004 Feb;3(2):145-8.

32. Wang C, Li Z, Fu M, Bouras T, Pestell RG. Signal transduction mediated by cyclin D1: from mitogens

to cell proliferation: a molecular target with therapeutic potential. Cancer Treat Res. 2004;119:217-37.

33. Arnold A, Papanikolaou A. Cyclin D1 in breast cancer pathogenesis. J Clin Oncol. 2005 J un

20;23(18):4215-24.

34. Aaltomaa S, Kärjä V, Lipponen P, Isotalo T, Kankkunen JP, Talja M, Mokka R. Expression of Ki-67,

cyclin D1 and apoptosis markers correlated with survival in prostate cancer patients treated by radical

prostatectomy. Anticancer Res. 2006 Nov-Dec;26(6C):4873-8.

35. Casimiro M, Rodriguez O, Pootrakul L, Aventian M, Lushina N, Cromelin C, Ferzli G, Johnson K,

Fricke S, Diba F, Kallakury B, Ohanyerenwa C, Chen M, Ostrowski M, Hung MC, Rabbani SA, Datar

22

R, Cote R, Pestell R, Albanese C. ErbB-2 induces the cyclin D1 gene in prostate epithelial cells in vitro

and in vivo. Cancer Res. 2007 May 1;67(9):4364-72.

36. Revelos K, Petraki C, Gregorakis A, Scorilas A, Papanastasiou P, Koutsilieris M. Immunohistochemical

expression of Bcl2 is an independent predictor of time-to-biochemical failure in patients with clinically

localized prostate cancer following radical prostatectomy. Anticancer Res. 2005 Jul-Aug;25(4):3123-33.

23

Legends to Figures

Figure 1. Bmi-1 protein levels in (A) normal prostate epithelial cells, CaP cells and (B) human prostatic

tumor tissues. (A) Expression of Bmi-1 protein in NHPE, RWPE1 and prostate cancer cells LNCaP, DU145

and PC-3 by western blotting. Equal loading of protein was confirmed by stripping the blots and reprobing with

β-actin antibody. The histogram indicates the relative density of the bands normalized to β-actin. Representative

data for five experiments are shown here. Each bar represents mean of relative densities ± S.E. NS, non-

significant. (B) Immunostaining for matriptase in representative specimens of CaP specimens of tumor stages I-

III and non-neoplastic regions of prostatic specimens of CaP patients. CaP specimens were assigned tumor

grades on the basis of Gleason pattern and Gleason score as described in Materials and Methods.

Immunoreactive Bmi-1 protein was observed in a coarsely granular pattern in cell cytoplasms of epithelial cells

of grade 1, grade 2 and grade 3 prostatic adenocarcinoma. There was minimal staining of occasional stromal

cells in non-neoplastic regions. Bmi-1 expression was weak in normal and moderate to strong in advanced CaP

specimens. Arrows indicate staining for Bmi-1 in cancer regions. Magnification X 40.

Figure. 2. Effect of Bmi-1-siRNA and pbabe-Bmi-1 plasmid transfection on the expression levels of Bmi-1

protein in CaP cells: (A) Immunoblots represent the effect of Bmi-1-siRNA transfection on the expression

level of Bmi-1 protein in LNCaP, DU145 and PC-3 cells. Cells were transfected with 100 nM of siRNA

directed against Bmi-1 and scrambled siRNA (100 nM). Cells were harvested at 24 and 36 h post transfection.

Cell lysates were prepared as described under Materials and methods. The expression level of Bmi-1 was

determined by western blot analysis. Equal loading was confirmed by stripping immunoblots and reprobing

them for β-actin. (B) Immunoblots represent the effect of pbabe-Bmi-1 plasmid transfection on the expression

level of Bmi-1 protein in LNCaP, DU145 and PC-3 cells at 36 h post transfection. Cells were transfected with 2

mg of pbabe-Bmi-1 plasmid and equal amount of vector (pbabe-puro). Cells were harvested at 24 and 36 h post

transfection. Cell lysates were prepared as described under Materials and methods. The expression level of

Bmi-1 was determined by western blot analysis. Equal loading was confirmed by stripping immunoblots and

reprobing them for β-actin.

24

Figure 3. Effect of Bmi-1-knockdown and Bmi-1 over-expression on the growth and viability of CaP cells:

(A) The histogram represents the % viability of human CaP cells LNCaP, DU145 and PC-3 at 72 h post-

transfection as measured by MTT assay. (B) The histogram represents the % viability of human CaP cells

LNCaP, DU145 and PC-3 at 36 h post-transfection as measured by MTT assay.

Figure 4. Effect of Bmi-1-knockdown and Bmi-1 over-expression on the rate of proliferation of CaP cells:

(A)Histogram showing rate of 3[H]thymidine uptake in Bmi-1-silenced CaP cells, LNCaP, DU145 and PC-3.

Cells were transfected with Bmi-1-siRNA (100nM). Control cells were transfected with scrambled siRNA

(100nM). Cells were incubated for 36h, t he last 16 h of which were in the presence of [3H]thymidine (0.5

μCi/ml). Each bar represents mean ± SE of three independent experiments. *indicates p<0.05. (B) Histogram

showing rate of 3[H ]thymidine uptake in Bmi-1-overexpressed CaP cells, LNCaP, DU145 and PC-3. Cells were

transfected with pbabe-Bmi-1(2 µg). Control cells were transfected with pbabe vector alone (2 µg). Cells were

incubated for 36h, t he last 16 h of which were in the presence of [3H] thymidine (0.5 μCi/ml). Each bar

represents mean ± SE of three independent experiments. *indicates p<0.05.

Figure 5. Effect of Bmi-1-knockdown and Bmi-1 over-expression on the clonogenic potential of CaP cells.

(A)Histogram showing number of colonies formed by Bmi-1-silenced LNCaP, DU145 and PC-3 cells. Bmi-1-

silenced cells were seeded in agarose and incubated at 37°C as described under Materials and methods. After 10

days of incubation, the cells were stained with crystal violet/methanol and colonies were counted. Each bar in

the histogram represents mean ± S.E., * indicates p<0.05. All experiments were repeated three times with

similar results. (B)Histogram showing number of colonies formed by Bmi-1-overexpressing LNCaP, DU145

and PC-3 cells. Bmi-1-overexpressing CaP cells were seeded in agarose and incubated at 37°C as described

under Materials and methods. After 10 days of incubation, the cells were stained with crystal violet/methanol

and colonies were counted. Each bar in the histogram represents mean ± S.E., * i ndicates p<0.05. All

experiments were repeated three times with similar results.

Figure 6. Effect of Bmi-1-knockdown and Bmi-1 over-expression on the expression level of prominent

proliferation-associated proteins in CaP cells. (A) Immunoblots represent the effect of Bmi-1-knockdown on

25

the expression level of Cyclin D1, Bcl-2 and p16 proteins in LNCaP, DU145 and PC-3 cells. Cells were

transfected with 100 n M of siRNA directed against Bmi-1 and scrambled siRNA (100 nM). Cells were

harvested at 36 h post transfection. Cell lysates were prepared as described under Materials and methods. The

expression level of Cyclin D1, Bcl-2 and p16 w ere determined by western blot analysis. Equal loading was

confirmed by stripping immunoblots and reprobing them for β-actin. (B) Immunoblots represent the effect of

Bmi-1-overexpression on the expression level of Cyclin D1, Bcl-2 and p16 proteins in LNCaP, DU145 and PC-

3 cells. Cells were transfected with 2 mg of pbabe-Bmi-1 plasmid and equal amount of vector (pbabe-puro).

The expression levels of proteins were determined by western blot analysis. Equal loading was confirmed by

probing the immunoblots for β-actin. The immunoblots shown here are representative of three independent

experiments with similar results.

Figure 7. Effect of Bmi-1-knockdown and Bmi-1 over-expression on the transcriptional activation TCF-

responsive element in CaP cells. Histogram represents the effect of Bmi-1 over-expression on the

transcriptional activation of TCF responsive element (marker of Wnt/β-catenin signaling) in LNCaP, DU145

and PC-3 cells. CaP cells were transfected with pTK-TCF-Luc (pTopFlash)-constructs. pFopFlash and Renilla

luciferase were used as negative and internal control respectively. For controls, the same amount of empty

vectors, were transfected in cells. The transcriptional activity was measured in terms of luciferase activity as

described under Materials and methods. Relative luciferase activity was calculated with the values from vector

alone group.

Figure 8. Effect of Bmi-1 over-expression on the rate of replication or proliferation of normal prostate

epithelial cells (PrEC): Micrographs showing the morphology of Bmi-1 overexpressing PrEc cells. PrEC cells

were transfected with 2 mg of pbabe-Bmi-1 plasmid and equal amount of vector (pbabe-puro). Confluent dishes

containing Bmi-1-overexpressing cells and vector-transfected PrEC cells were split or seeded after every 36 h.

(upper Panel) Cell splitting or seeding continued for 4 passages or replication cycles in pbabe-transfected PrEC

cells and did not duplicate after 4 passages and entered into senescence phase.(Lower Panel) Cell splitting or

seeding continued for 8 passages or replication cycles in pbabe-transfected PrEC cells and cell replication

26

continued upto 4 passages. Since this was a transient transfection and the overexpression effect of Bmi-1 lasted

upto 8th passage only. Cells entered into senescence phase at 9th passage. Inset regions (400X) showing cells

with senescent morphology features of live cells such as globular shape.

Figure 9. Effect of Sonic Hedgehog signaling inhibition on the growth of CaP cells exhibiting varied Bmi-

1 expression levels. Histogram represents the effect of Cyclopamine (SHH inhibitor) treatment for 12 h on the

viability of Bmi-1-suppressed and Bmi-1-overexpressing LNCaP cells. (A) To achieve Bmi-1 knockdown, cells

were transfected with 100 nM of siRNA directed against Bmi-1 and scrambled siRNA (100 nM). For controls,

the same amount of scrambled siRNA were transfected in cells. At 24 h post transfection, cells were treated

with cyclopamine in fresh media. Control cells were treated with DMSO (vehicle alone). After 12 h incubation

of cells with cyclopamine or vehicle alone, cell viability was measured by employing MTT assay. (B) To

achieve Bmi-1 overexpression, cells were transfected with 2 mg of pbabe-Bmi-1 plasmid and equal amount of

vector (pbabe-puro). For controls, the same amount of empty vectors, were transfected in cells. At 24 h post

transfection, cells were treated with cyclopamine in fresh media. Control cells were treated with DMSO

(vehicle alone). After 12 h incubation of cells with cyclopamine or vehicle alone, cell viability was measured by

employing MTT assay.

Figure 10. Effect of Cyclopamine treatment on the expression level of Bcl-2 in Bmi-1-suppressed and

Bmi-1-overexpressing CaP cells. (A) Immunoblots represent the effect of cyclopamine treatment on the

expression level of Bcl-2 protein in LNCaP cells. Cells were transfected with 100 nM of siRNA directed against

Bmi-1 and scrambled siRNA (100 nM). For controls, the same amount of scrambled siRNA was transfected in

cells. At 24 h post transfection, cells were treated with cyclopamine in fresh media. Control cells were treated

with DMSO (vehicle alone). After 12h incubation with either cyclopamine or vehicle alone (DMSO), cells were

harvested. Cell lysates were prepared as described under materials and methods. The expression level of Bcl-2

protein was determined by western blot analysis. Equal loading was confirmed by stripping immunoblots and

reprobing them for β-actin. (B) Immunoblots represent the effect of cyclopamine treatment on the expression

level of Bcl-2 protein in Bmi-1 overexpressing LNCaP cells. To achieve Bmi-1 over-expression, cells were

27

transfected with 2 mg of pbabe-Bmi-1 plasmid. For controls, the same amount of empty vector (pbabe-puro)

were transfected in cells. At 24 h post transfection, cells were treated with cyclopamine in fresh media. Control

cells were treated with DMSO (vehicle alone). After 12 h incubation of cells with cyclopamine or vehicle alone

cells were harvested. Cell lysates were prepared as described under materials and methods. The expression level

of Bcl-2 protein level was determined by western blot analysis. Equal loading was confirmed by stripping

immunoblots and reprobing them for β-actin.

Figure 11. Analysis of Bcl-2 promoter region for TCF transcriptional binding sites. Bcl-2 promoter region

was analyzed for binding sites for transcriptional factors by employing a web-based TESS-analysis program.

Multiple binding sites for TCF transcriptional factor (key component of Wnt signaling) on promoter region of

Bcl-2 gene were identified and are presented in the figure.

Figure 12. Effect of TCF-knockdown on the transcriptional activation of Bcl-2. Histogram represents the

effect of TCF-knockdown on the transcriptional activation of Bcl-2 in LNCaP and PC-3 cells. CaP cells were

transfected with pGL3-Bcl-2-Luc construct. For controls, the same amount of empty vector (pGL3) were

transfected in cells. Renilla luciferase was used as internal control. The transcriptional activity was measured in

terms of luciferase activity as described under materials and methods. Relative luciferase activity was calculated

with the values from vector alone group. The data is presented as relative luciferase units (RLU).

Figure 13. Effect of Bmi-1-introduction on Bcl-2 promoter activity in TCF-dominant negative LNCaP

cells. Histogram represents the effect of re-introduction of Bmi-1 on the transcriptional activation of Bcl-2 in

tcf-dominant negative (dn) LNCaP cells. CaP cells were co-transfected with Tcf-dominant negative vector and

pGL3-Bcl-2-Luc construct. Bcl-2 transcriptional activation was measured in these cells. Tcf-dn LNCaP cells

were co-transfected with pbabe-Bmi-1 and pGL3-Bcl-2-luc plasmid. Bcl-2 transcriptional activation was

measured in these cells. For controls, the same amount of empty vector (pGL3) were transfected in cells.

Renilla luciferase was used as internal control. Relative luciferase activity was calculated with the values from

vector alone group. The data is presented as relative luciferase units (RLU).

28

Figure 14. Effect of Bmi-1-introduction on Bcl-2 promoter activity in LNCaP cells pre-treated with SHH

inhibitor Cyclopamine. Histogram represents the effect of Bmi-1 introduction on the transcriptional activation

of Bcl-2 in LNCaP cells pretreated with SHH inhibitor Cyclopamine. LNCaP cells were transfected with pGL3-

Bcl-2-Luc construct. 24h after transfection, LNCaP cells were treated with Cyclopamine for 24 h. After 24 h of

cyclopamine treatment, Bcl-2 promoter activity in cells was measured. LNCaP cells were transfected with

pGL3-Bcl-2-Luc construct. For controls, the same amount of empty vector (pGL3) and were transfected in cells

treated with DMSO (vehicle control). Renilla luciferase was used as internal control. In another set of

experiment (under similar culture conditions) LNCaP cells were treated with Cyclopamine for 24 h. After 24 h,

cyclopamine treated cells were washed with PBS and were co-transfected with pbabe-Bmi-1 and pGL3-Bcl2-

luc. Renilla luciferase was used as internal control. The Bcl-2 transcriptional activity was measured. Relative

luciferase activity was calculated with the values from vector alone group. The data is presented as relative

luciferase units (RLU).

Figure 15. Effect of pGeneCLIP-Bmi-1-shRNA and pbabe-Bmi-1 plasmid transfection on the expression

levels of Bmi-1 and Bcl-2 proteins in stable PC-3 CaP cells: (A) Bmi-1 expression was knocked down or

overexpressed in stably transfected PC-3 cell line. Immunoblot analysis shows that Bmi-1 is increased (Lane 1

and 2 for BO) in much higher levels in the stable clones’ transfected with pbabe-Bmi-1 after selection compared

to control PC-3 cells. Similarly, Bmi-1 is reduced to very low levels in the cells stably transfected with the short

hairpin RNA (shRNA)-Bmi-1 plasmid (Lane 1 and 2 for BS) compared to control PC-3 cells. Cell lysates were

prepared as described under Materials and methods. The expression level of Bmi-1 was determined by western

blot analysis. Equal loading was confirmed by stripping immunoblots and reprobing them for β-actin. (B)

Immunoblot represents the effect of Bmi-1 silencing and overexpression on t he expression level of Bcl-2

protein in stable PC-3 cells. Cell lysates were prepared as described under Materials and methods. The

expression level of Bcl-2 was determined by western blot analysis. Equal loading was confirmed by stripping

immunoblots and reprobing them for β-actin. (C) Micrographs showing the morphology of Bmi-1

29

overexpressing, silencing and empty vector transfected stable PC-3 cells. Confluent dishes containing Bmi-1-

overexpressing, -silencing and empty vector transfected PC-3 cells. Magnification at 100X.

Figure 16. Effect of Bmi-1 overexpression on PC-3 tumor growth in athymic nude mice. (A) The graphical

representation of data showing the effect of Bmi-1 overexpression on growth of tumors from PC-3 cells

implanted in athymic nude mice. The growth was measured in terms of average volume of tumors as a function

of time. Data is represented as mean ± SE, * i ndicates p< 0.05 from the control group (B) The graphical

representation of the data depicting the number of mice remaining with tumor volumes <1000 mm3 after Bmi-1

overexpression for indicated weeks. The growth of tumor volume was shown to higher in nude mice bearing a

pbabe-Bmi-1 stable PC-3 cell xenograft as compared to empty vector transfected PC-3 stable cells. PC-3-BO

represents pbabe-Bmi-1 transfected PC-3 stable cells (Bmi-1 overexpressing). The details are described under

Materials and Methods.

Figure 17. Effect of Bmi-1 suppression on PC-3 tumor growth in athymic nude mice. (A) The graphical

representation of data showing the effect of Bmi-1 suppression on growth of tumors from PC-3 cells implanted

in athymic nude mice. The growth was measured in terms of average volume of tumors as a function of time.

Data is represented as mean ± SE, * indicates p< 0.05 from the control group (B) The graphical representation

of the data depicting the number of mice remaining with tumor volumes <1000 mm3 after Bmi-1 suppression

for indicated weeks. The growth of tumor volume was shown to slow of nude mice bearing a pGeneCLIP-Bmi-

1-shRNA stable PC-3 cell xenograft as compared to empty vector transfected PC-3 cells. PC-3-BS represents

pGeneCLIP-Bmi-1-shRNA transfected PC-3 stable cells (Bmi-1 silenced). The details are described under

Materials and Methods.

last

ic

(B)Bmi-1

β ti

(A)

Non

-neo

p

NHPE RWPE1 LNCaP PC-3 DU-145

β-actin

age

II C

aPSt

a

Figure 1

CaP

Stag

e III

Cge

IV C

aPSt

ag

30

Bmi-1P

(B)(A)

aP Bmi-1

5 Bmi-1

Bmi-1

β-actinLNC

aP

Bmi-145LN

Ca

β-actin

Bmi 1

Bmi-13D

U14

Bmi 1

β-actinβ-actin

C-3

DU

14

Bmi-1

β-actin

pbabevector

pbabeBmi-1

PC-

ControlsiRNA

24 36 hBmi-1 siRNA

PC β-actin

Figure 2

31

2 h

Scrabmbled siRNA

(A)

80

100120

wth

afte

r 72 Scrabmbled siRNA

Bmi-1-siRNA

020

4060

% C

ell G

row

0% LNCaP DU145 PC-3

100

36 h pbabe-puro

pbabe-Bmi-1

(B)

40

60

80

row

th a

fter 3

pbabe-Bmi-1

0

20

40

% C

ell G

r

LNCaP DU145 PC-3Figure 3

32

(A)

ne u

ptak

e N

A x

102 )

300

400

500Control siRNABmi-1 siRNA

3 [H

]thym

idin

(cpm

/μgD

N0

100

200

300

*

**

0DU145 PC-3LNCaP

600

pbabe vectorpbabe-Bmi-1

*ke

2 )

(B)

200300400500

***

ymid

ine

upta

km

/μgD

NA

x 10

2

Figure 4

0100200

DU145 PC-3LNCaP

3 [H

]thy

(cpm

Figure 4

33

100125150

mbe

r of

s/fie

ld

Control siRNABmi-1 siRNA

(A)

0255075

Aver

age

num

Col

onie

sDU145 PC-3LNCaP

200250

DU145 PC-3LNCaP

ber o

fel

d

pbabe vectorpbabe-Bmi-1

(B)

050

100150200

Aver

age

num

bC

olon

ies/

fie

0

Figure 5

A DU145 PC-3LNCaP

34

Gene Bank Accession Number

Gene Description Gene status after Bmi-1-siRNA transfection

Table 1. Effect of Bmi-1 knockdown on Expression of Proliferation-Associated Genes in CaP Cells

NM_053-56NM_000075NM_000657NM_005163NM_004570NM 002658

Cyclin D1Cyclin-dependent kinase 4 (Cdk4)Bcl-2Akt1PI3K (catalytic gamma peptide)U ki l i ti t

Down-regulated **Down-regulated **Down-regulated **Down-regulated *Down-regulated *D l t d **NM_002658

NM_001005862NM_005417NM_002467NM_002228NM 005343

Urokinase plasminogen activatorErbb2Srcc-mycc-JunH-Ras

Down-regulated **Down-regulated **Down-regulated *Down-regulated **Down-regulated *Down-regulated *_

NM_001033756NM_0004994NM_003998NM_001530NM_000875NM 0008756

Vascular endothelial growth factor (VEGF)Matrix metalloproteinase -9 (MMP-9)Nuclear factor kappa B 1(NFκB1)HIF-1Insulin growth factor 1(IGF1)Insulin growth factor 1(IGF2)

gDown-regulated **Down-regulated **Down-regulated *Down-regulated *Down-regulated **Down regulated **NM_0008756

NM_000586NM_013430NM_006254NM_005400NM_058195

Insulin growth factor 1(IGF2)Interlukin 2(IL-2)GGT2Protein kinase C delta (PKC δ)Protein kinase C epsilon (PKC ε)p16/INK4

Down-regulated **Down-regulated **Down-regulated *Down-regulated *Down-regulated*Up-regulated **_

NM_004936NM_000076NM_006016NM_016279NM_000362NM 173206

pp15p57CD164Cadherin-9TIMP-3PIAS2

p gUp-regulated **Up-regulated **Up-regulated *Up-regulated *Up-regulated *Up regulated **NM_173206 PIAS2 Up-regulated

* Represents 2- 5 fold and ** represents more than 5 fold.

35

Cyclin D1

LNCaP Cells(A) DU145 Cells PC-3 CellsCyclin D1Cyclin D1Cyclin D1

Bcl-2 Bcl-2Bcl-2

yCyclin D1

p16 p16 p16

β-actin

controlsiRNA

Bmi-1siRNA

controlsiRNA

Bmi-1siRNA

controlsiRNA

Bmi-1siRNA

β-actinβ-actin

Cyclin D1

DU145 Cells PC-3 Cells

Cyclin D1Cyclin D1

LNCaP Cells(B)

β-actin β-actinβ-actin

Bcl-2 Bcl-2Bcl-2

p16 p16 p16

pbabevector

pbabeBmi-1

β β actin

pbabevector

pbabeBmi-1

Figure 6

pbabevector

pbabeBmi-1

β-actin

36

20

nt eras

e )

Tcf-luc + pbabe vectorTcf luc + pbabe-Bmi-1

10

15

onsi

ve e

lem

enR

elat

ive

Luci

fey

in fo

ld u

nits

Tcf luc + pbabe Bmi 1

0

5

Tcf-r

esp

activ

atio

n ( R

Act

ivity

LNCaP DU145 PC-3LNCaP DU145 PC 3

Figure 7

37

(400

1st 4th cyclepbabe-transfected PrEC Cells

0 x)1st 3th 5th 8th cycle

(400 x)

1st 3th 5th 8th cyclepbabe-Bmi-1-transfected PrEC Cells

Figure 8

38

50

75

100(B)

ell V

iabi

lity(A)

50

75

100

Cel

l Via

bilit

y

0

25% C

pbabe-Bmi-1 - - - + +pbabe vector - + + - -

Cyclopamine + +

0

25

Bmi-1 siRNA - - - + +scrambled siRNA - + + - -

Cyclopamine - - + - +

% C

*

Cyclopamine - - + - +DMSO alone + - - - -

Figure 9

Cyclopamine + +DMSO alone + - - - -

39

Bcl2 Bcl2

β-actin β-actin

(A) (B)

Bmi-1 siRNA - - + +scrambled siRNA + + - -

Cyclopamine - + - +

pbabe-Bmi-1 - - + +pbabe vector + + - -

Cyclopamine - + - +

Figure 10

40

(TESS Analysis)

Analysis of Promoter Region of Bcl-2 Gene

(TESS Analysis)

Length of Sequence for TCF binding on Bcl-2 promoter : 5Binding sites for

transcriptional factor TCF (T00999-T1001)

Bcl-2Bcl 2

1.98 kbp

Positions (kbp) 19/20 39 162 257 521 547 592 601 663 692 1058 1075 1081 1099 1152 1292 1393 1633 1814 1909

Figure 11

41

0 025

0.03

vity

(A)0.0004ity

LNCaP cells PC-3 cells(B)

0.015

0.02

0.025

ucife

rase

act

iv(B

cl-2

)

0.00020.000250.0003

0.00035

cife

rase

act

ivi

Bcl

-2)

*

0

0.005

0.01

Rel

ativ

e Lu

00.000050.0001

0.00015

Rel

ativ

e Lu

c (B

*

scrambledshRNA

TCF shRNA

Figure 12

scrambledshRNA

TCF shRNA

Figure 12

42

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

Bcl-2

pro

mot

er a

ctiv

atio

nRe

lativ

e Lu

cife

rase

Uni

ts (R

LU)

Empty vector

Tcf-dominantnegative plasmid

Tcf-dn plasmid+

pbabe vector

Tcf-dn plasmid+

pbabe Bmi-1

*

*

Figure 13

Re-introduction of Bmi-1in Tcf-dn LNCaP cells

4143

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

Bcl-2

pro

mot

er a

ctiv

atio

nRe

lativ

e Lu

cife

rase

Uni

ts (R

LU)

DMSO(VehicleControl)

Cyclopamine Cyclopamine+

pbabe vector

Cyclopamine+

pbabe Bmi-1

Figure 14

Re-introduction of Bmi-1in LNCaP cells

after 48 h of Cyclopaminetreatment

44

Figure 15

PC.3 (stable)

Bm~1r--- I 13-actin I I

1 2 1 2 PC-3 A BO BS

Bcl-2 1 I 13-actin I

BO BS Cont

B

45

Figure 16

1200

1000

~800 E

i 6001

0

~ 400 0

~ 200 1

o.--------------------------------------A

8

0 ., 8 9 10

0+-----------------~~----~~----~~--0 1 234 -5678910 Weeks Post-inoculation

46

Figure 17

A

B

1

"10

2 3 4 5 6 7 8 W eeks ~{Post-inoculation)

9 10

0+---------------------~--~------------~~---0 1 2 3 4 5 6 7 8

47